IEEE Conference Paper Template

Transcription

International Journal of Science, Technology & Management
Volume No.04, Issue No. 05, May 2015
www.ijstm.com
ISSN (online): 2394-1537
IMAGE PROCESSING BASED LANGUAGE
CONVERTER FOR DEAF AND DUMB
S.N.Boraste1, K.J.Mahajan2
1
PG Student, Department of Electronics and Telecommunication Engineering KCT’s Late G. N.
Sapkal College of Engineering, Anjaneri, Nashik, (India)
Department of Electronics and Telecommunication Engineering KCT’s Late G. N.
2
Sapkal College of Engineering, Anjaneri, Nashik, (India)
ABSTRACT
This application helps the deaf and dumb person to communicate with the rest of the world using sign language.
Suitable existing methods are integrated in this application. Computer recognition of sign language is an
important research problem for enabling communication with hearing impaired people. The Computer based
intelligent system will enable deaf & dumb people significantly to communicate with all other people using their
natural hand gestures.
Index Terms: This Application Helps The Deaf And Dumb Person To Communicate With The
Rest of The World Using Sign Language
I. INTRODUCTION
Deaf and Dumb people are usually deprived of normal communication with other people in the society. It has
been observed that they find it really difficult at times to interact with normal people with their gestures, as only
a very few of those are recognized by most people. Since people with hearing impairment or deaf people cannot
talk like normal people so they have to depend on some sort of visual communication in most of the time. Sign
Language is the primary means of communication in the deaf and dumb community. As like any other language
it has also got grammar and vocabulary but uses visual modality for exchanging information. The problem arises
when dumb or deaf people try to express themselves to other people with the help of these sign language
grammars. This is because normal people are usually unaware of these grammars. As a result it has been seen
that communication of a dumb person are only limited within his/her family or the deaf community. The
importance of sign language is emphasized by the growing public approval and funds for international project.
At this age of Technology the demand for a computer based system is highly demanding for the dumb
community. However, researchers have been attacking the problem for quite some time now and the results are
showing some promise. Interesting technologies are being developed for speech recognition but no real
commercial product for sign recognition is actually there in the current market. So, to take this field of research
to another higher level this project was studied and carried out. The basic objective of this research was to
develop a computer based intelligent system that will enable dumb people significantly to communicate with all
other people using their natural hand gestures. The idea consisted of designing and building up an intelligent
109 | P a g e
www.ijstm.com
system using image processing, data mining and artificial intelligence concepts to take visual inputs of sign
language‟s hand gestures and generate easily recognizable form of outputs in the form of text & Voice.
II. RELATED WORK
The literature survey is carried out as a part of the project work. It has provided review of the past research about
image processing based language converter and other researchers. The past research effort will properly guide to
justify the scope and direction of the present effort.
This chapter reviews the literature on sign language & its application in various national and international
journals. Not many Researches have been carried out in this particular field, especially in Binary Sign Language
Recognition. Few researches have been done on this issue though and some of them are still operational, but
nobody was able to provide a full fledged solution to the problem. Christopher Lee and Yangsheng Xu
developed a glove-based gesture recognition system that was able to recognize 14 of the letters from the hand
alphabet, learn new gestures and able to update the model of each gesture in the system in online mode, with a
rate of 10Hz. Over the years advanced glove devices have been designed such as the Sayre Glove, Dexterous
Hand Master and Power Glove [1].
The most successful commercially available glove is by far the VPL Data Glove [2]. It was developed by
Zimmerman during the 1970‟s. It is based upon patented optical fiber sensors along the back of the fingers. Starner and Pentland developed a glove-environment system capable of recognizing 40 signs from the American
Sign Language (ASL) with a rate of 5Hz.
Another research is by Hyeon-Kyu Lee and Jin H. Kim presented work on real-time hand-gesture recognition
using HMM (Hidden Markov Model). Kjeldsen and Kendersi devised a technique for doing skin-tone
segmentation in HSV space, based on the premise that skin tone in images occupies a connected volume in HSV
space. They further developed a system which used a back-propagation neural network to recognize gestures
from the segmented hand images [1].
Etsuko Ueda and Yoshio Matsumoto presented a novel technique a hand-pose estimation that can be used for
vision-based human interfaces, in this method, the hand regions are extracted from multiple images obtained by a
multi viewpoint camera system, and constructing the “voxel Model”[6] . Hand pose is estimated. Chan Wah Ng,
Surendra Ranganath presented a hand gesture recognition system, they used image furrier descriptor as their
prime feature and classified with the help of RBF network. Their system‟s overall performance was 90.9%.
Claudia Nolker and Helge Ritter presented a hand gesture recognition modal based on recognition of finger tips,
in their approach they find full identification of all finger joint angles and based on that a 3D modal of hand is
prepared and using neural network.
Their complex spatial grammars are remarkably different from the grammars of spoken languages [1], [2]. of
sign languages, such as ASL (American Sign Language), BSL (British Sign Language), Auslan (Australian Sign
Language) and LIBRAS (Brazilian Sign Language) [1], are in use around the world and are at the cores of local
deaf cultures. Unfortunately, these languages are barely known outside of the deaf community.
Depth information makes the task of segmenting the hand from the background much easier. Depth information
can be used to improve the segmentation process, as used in [4], [5], [6], [7]. Recently, depth cameras have
raised a great interest in vision computer community due to their success in many applications, such as pose
estimation [8], [9], tracking [10], object recognition [10], etc. Depth cameras were also used for hand gesture
110 | P a g e
www.ijstm.com
recognition [11], [12], [13]. Uebersax et al. [12] present a system for recognizing letter and finger spelled words.
Pugeault & Bowden [11] use a Microsoft.
Kinect TM device to collect RGB and depth images. They extracted features using Gabor filters and then a
Random Forest predicts the letters from the American Sign Language (ASL) finger spelling alphabet. Issacs &
Foo [14] proposed an ASL finger spelling recognition system based on neural networks applied to wavelets
features. Bergh & Van Gool [15] propose a method based on a concatenation of depth and colorsegmented
images, using a combination of Haar wavelets and neural networks for 6 hand poses recognition of a single user.
Several techniques have been reported on gesture recognition which includes skin segmentation using color pixel
classification [1], region growing by exemplar-based hand segmentation under complex background [2],
Parametric Hidden Markov models for gesture recognition [3], statistical database comparison method [4],
accelerometer-based gesture recognition system [5], orientation histograms for gesture recognition [6], Finger
Detection for Sign Language Recognition[7] etc. Most of the gesture recognition systems use special devices
like hand glove [11]. The gloves get connected to the computers using a lot of cables. So these devices are
cumbersome and expensive. In order to overcome these difficulties, alternatively vision-based approaches
involving camera and image processing for recognizing gestures are being explored. In 2005, there were two
significant proposals [1, 2] for treating color to grayscale conversion as an optimization problem involving either
local contrasts between pixels [1] or global contrasts between colors [2]. However, unconstrained optimization
over thousands of variables [1] or constrained optimization over hundreds of variables [2] is a slow process
prone to local minima. It can prove impractical in applications that demand real-time, automated performance
with mathematical guarantees on the results. Our algorithm satisfies these requirements without resorting to
numerical optimization. Hence, our algorithm could easily be embedded in the driver software that renders color
images on a grayscale printer. Also, its design is suitable for hardware implementation. Our running time is
linear in the number of pixels, unlike the Gooch et al. algorithm [1], and is independent of the number of colors
in the image, unlike the Rasche et al. algorithm [2]. Owing to their complexity, the optimization algorithms [1,2]
offer few guarantees to aid in the interpretation of their results. For instance, they can fail to map pure white in
the color image to pure white in the grayscale image, causing potential problems for pictures with white
backgrounds. While our algorithm does not purport to model human visual adaptation, it ensures that converting
color to grayscale leaves achromatic pixels unchanged. The closest precedent for our approach to global color to
grayscale conversion is the classic work of Strickland et al. [3] on saturation feedback in unsharp masking, a
local color image enhancement technique used to sharpen images. However, our method is designed to take
advantage of both hue and saturation differences when augmenting the luminance channel.
III. SYSTEM DEVELOPMENT
The goal of proposed method is to convert RGB to text massage. Fig.1 shows the overall idea of proposed
system The system consists of 3 modules. Image is captured through the webcam. The camera is mounted on top
of system facing towards the wall with neutral background. Firstly, the captured Colored image is converted into
the gray scale image which intern converted into the binary form. Coordinates of captured image is calculated
with respect to X and Y coordinates. The calculated coordinates are then stored into the database in the form of
template. The templates of newly created coordinates are compared with the existing one. If comparison leads to
success then the same will be converted into audio and textual form. The system works in two different mode i.e.
111 | P a g e
www.ijstm.com
training mode and operational mode. Training mode is part of machine learning where we are training our
system to accomplish the task for which it is implemented i.e. Alphabet Recognition.
Figure 1: System Block Diagram
IV. EXPERIMENTAL RESULTS
The experimental results of our project are as shown below:Up to explain how system works theoretically. How RGB colors are detected & and how convert RGB to binary.
As per experimental results explain how Matching template Output of system in text and audio.
The algorithm is implemented using C# & .net using various real time and standard images. The real time images
are captured from web camera.The experimental results show the robustness of algorithm in both real time and
standard images even in case of rotation, missing data.The main motive of this work is to Convert RGB to binary
in form of audio & text.
4.1 RGB Color Recognition
The above Fig.2 indicates the red image captured using web camera & is the target image for binary conversion.
Figure 2: Captured Red Image Using Web Camera
4.2 Conversion of RGB Color to Pixel data
After capturing all above images using webcam, different pixel values calculated as shown in below image.
112 | P a g e
www.ijstm.com
Figure 3: Captured Pixel Values for Red Image Using Web Camera
4.3 Color image to Binary image conversion\
For above Pixel values below is the binary image for red color.
Figure 4: Captured Binary Image for Red Image Using Web Camera
4.4 Coordinate Mapping
After getting the marker pixels that are now highlighted as a white color pixels, coordinates of that area for each
color is generated. The newly generated coordinates are the compared with the stored coordinates in the database
for the purpose of output generation using pattern matching technique explained in the next section.
Figure 5: Co-ordinate Mapping
In this method, the input image after processing is set to the pixel values (3) of each color
to be used such
as Red_new (Rx, Ry), Green_new (Gx, Gy), Blue_new (Bx, By), Purple_new (Px, Py), Parrot Green _new
(PGx, PGy). Pixel values comprise of the minimum and maximum values of each color pixel or can be called as
co-ordinates. The generated values of these co-ordinates will be then compared with the values stored in the
templates stored in the database. To obtain these values the general idea is firstly to find the area of each color
pixel and the coordinates (Yx,Yy) by using the equation: Area= count number of white pixels obtained by
Thresholding.
113 | P a g e
www.ijstm.com
4.5 Alphabet Recognition
Following table 1 shows the values assigned to each finger. Binary Alphabet calculation: It is possible to display
total (25−1) i.e.32 gestures using the fingers of a single hand.
Table.1 Values assigned to Each Color
Table 2. Alphabets Code
4.6 RGB to Text Conversion
Putting pixel values to run the program for above different colors giving text massages as per above alphabate
codes. Below images are the example shown for Red color. While showing Red input as shown in below input
Fig 6 infront of webcam text form shows „a‟ alphabate as shown in below output image 4.8. In the similar way
deaf & dump can talk with other person using different colors. Below output image 4.9 is the example of one of
the statement.
114 | P a g e
Figure 6: Input Image
www.ijstm.com
Figure 7: Output Image
V. CONCLUSION
In this way see that
1. How system block diagram works theoretically.
2. How RGB detect .
3. How RGB to binary conversion work its result.
4. How Binary to text massage & audio result.
VI. ACKNOWLEDGMENT
Acknowledgment not just being a hollow formality, I take this opportunity to express whole-heardly, my sincere
thoughts to all those who were responsible for the successful completion of this dissertation.
With a deep sense of gratitude, I wish to express my sincere thanks to my guide,Prof. K.J.Mahajan E&TC
Engineering Department, without whom this dissertation would not have been possible. The confidence and
dynamism with which she guided the work requires no elaboration. With her profound knowledge, the timely
suggestions given and encouragement shown, made me more confident in successfully carrying out this project
work. The opportunities given by her have stimulated many of my interests and enabled me to gain more
exposure in the area of my work.
I am also thankful to Prof. Bagal S. B., Head of Electronics & Telecommunication Engineering Department, for
giving me an opportunity to carry out the dissertation work within the college premises and allowing me to
available the departmental facilities.
I express my thanks to Principal Dr. Balapgol B. S., for his continuous support. I am also thankful to all teaching
& non- teaching staff of Electronics & Telecommunication Department for their kind co-operation and guidance
for preparing and presenting this project work.
This acknowledgement would not be complete without the mention of all my colleagues to whom I am grateful
for all their help, support, interest and valuable hints.
Finally, I would like to thank my parents for giving their most patient support and encouragement to my work.
REFERENCES
[1]
Christopher Lee and Yangsheng Xu, “Online, interactive learning of gestures for human robot interfaces”
Carnegie Mellon University, the Robotics Institute, Pittsburgh, Pennsylvania, USA, 1996
[2]
Richard Watson, “Gesture recognition techniques”, Technical report, Trinity College, Department of
Computer Science, Dublin, July, Technical Report No. TCD-CS-93-11, 1993
115 | P a g e
[3]
www.ijstm.com
Ray Lockton, ”Hand Gesture Recognition using computer Vision”,4th year project report, Ballilol
College , Department of Engineering Science, Oxford University,2000
[4]
Son Lam Phung, Member, IEEE, Abdesselam Bouzerdoum, Sr.Member, IEEE, and Douglas Chai, Sr.
Member, IEEE, "Skin Segmentation Using Color Pixel Classification: Analysis and Comparison", IEEE
transactions on pattern analysis and machine intelligence, vol. 27, no. 1, january 2005
[5]
Ma De-yi, Chen Yi-min, Li Qi-ming, Huang Chen, Xu Sheng, "Region Growing by Exemplar-Based
Hand Segmentation under Complex Backgrounds"
[6]
Andrew D. Wilson, Student Member, IEEE Computer Society, and Aaron F. Bobick, Member, IEEE
Computer Society, “Parametric Hidden Markov Models for Gesture Recognition”, IEEE transactions on
pattern analysis and machine intelligence, vol. 21, no. 9,september 1999
[7]
Chance M. Glenn, Divya Mandloi, Kanthi Sarella, and Muhammed Lonon, "An Image Processing
Technique for the Translation of ASL Finger-Spelling to Digital Audio or Text"
[8]
Ahmad Akl, Student Member, IEEE, Chen Feng, Student member, IEEE, and Shahrokh Valaee, Senior
Member, IEEE, “A Novel Accelerometer-Based Gesture Recognition System”, IEEE transactions on
signal processing, Vol.59, No.12, December 2011.
[9]
Meenakshi Panwar, Hand Gesture Recognition System based on Shape parameters, International
Conference on Recent Advances in Computing and Software Systems, pp. 80-85, February 2012.
[10]
Meenakshi Panwar , P.S. Mehra, Hand Gesture Recognition for Human Computer Interaction, IEEE
International Conference on Image Information Processing, pp. 1-7, November 2011.
[11]
Rohit Verma, Ankit Dev, Vision based Hand Gesture Recognition Using Finite State Machines and Fuzzy
Logic, IEEE International Conference on Ultra Modern Technologies, pp 1-6, October 2009.
[12]
V. Frati and D. Prattichizzo, Using Kinect for hand tracking and rendering in wearable haptics,” in
Proceedings of the IEEE World Haptics Conference (WHC). IEEE, 2011, pp. 317–321.
[13]
Y. Li, “Hand gesture recognition using Kinect,” in Proceedings of the 3rd IEEE International Conference
on Software Engineering and Service Science (ICSESS). IEEE, 2012, pp. 196–199.
[14]
Z. Mo and U. Neumann, “Real-time hand pose recognition using lowresolutiondepth images,” in
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol.
2. IEEE, 2006, pp. 1499–1505.
[15]
N. Pugeault and R. Bowden, “Spelling it out: Real-time ASL fingerspelling recognition.” in Proceedings
of the IEEE International Conference on Computer Vision Workshops (ICCV Workshops). IEEE, 2011,
pp. 1114–1119.
[16]
D. Uebersax, J. Gall, M. V. den Bergh, and L. J. V. Gool, “Real-time sign language letter and word
recognition from depth data,” in Proceedings of the IEEE International Conference on Computer Vision
Workshops (ICCV Workshops), 2011, pp. 383–390.
[17] C. L. Novak and S. A. Shafer, “Color edge detection,” in Proc. DARPA Image Understanding Workshop,
1987, pp. 35–37. [18] R. Nevatia, “A color edge detector and its use in scenesegmentation,” IEEE Trans.
Syst., Man, Cybern., vol. SMC-7,no. 11, pp. 820–826.
[19]
Rafael C. Gonzalez, Richard E. Woods, Digital Image Processing, Second Edition.
[20]
A. Shiozaki, “Edge extraction using entropy operator,” Computer Vis.,Graph., Image Process., vol. 36,
no. 1, pp. 1–9, Oct. 1986.
116 | P a g e
[21]
www.ijstm.com
M. A. Ruzon and C. Tomasi, “Color edge detection with compass operator,” in Proc. IEEE Conf.
Computer Vision Pattern Recognition, Jun. 1999, vol. 2, pp. 160–166.
[22]
A. Cumani, “Edge detection in multispectral images,” CVGIP: Graph. Models Image Process, vol. 53, no.
1, pp. 40–51, Jan. 1991.
[23]
S. K. Naik and C. A. Murthy, “Standardization of Edge Magnitude in Color Images,” IEEE Trans. Image
processing, vol. 15, no. 9, Sept. 2006.
117 | P a g e

IEEE Conference Paper Template

Transcription

Similar documents

Region 10 Meeting Report Dhaka Bangladesh, March 7

Congratulations on having your paper accepted for - ICCCT-2014

Details on running

Call for Papers - Computer Science

Miroku`s Gesture - Phoebus Media Group

2015 Call for speakers - Princeton ACM / IEEE Computer Society

here - IEEE-RFID

Call for Papers PowerAfricaCFP_flyer

Call for Papers, Workshops, and Tutorials

Time: 6:00 to 8:00 PM Where: Shakey`s Pizza Parlor 12540 Valley